• Associate Professor: Biomedical, Electrical and Computer Engineering, Toronto Metropolitan University
• Principal Investigator: Image Analysis in Medicine Lab (IAMLAB)
• Affiliated Scientist: St. Michael's Hospital & iBEST
What inspired you to pursue using machine learning for medical images?
I love using mathematics, machine learning, and engineering theories to solve real problems for society, and I've been fascinated by medicine and biology since childhood. Focusing on machine learning for medical images allowed me to combine my passions while helping others. Since medical imaging plays such an important role in diagnosis and treatment, I couldn’t dream of a better way to use my skills.
What outcomes do you hope your research will eventually lead to?
My goal is to create AI tools for radiology and pathology images that clinicians can use to improve quality of care, turn-around-times, and access. AI tools for medical imaging can lead to more objective measures of disease, improved diagnostic accuracy, and reduced inter-rater disagreement. And these all lead to better patient management. I am building AI tools that augment neuroradiology and digital pathology workflows, and in the next few years I hope my algorithms are used to automatically populate imaging reports with images, statistics and measurements, as part of routine clinical care.
What excites you the most about the possibilities of AI in healthcare?
I am excited since it has the potential to change the way medicine is practiced. I like to think of AI tools for medical imaging as a software-based “toolbox” that arms clinicians with objective measurements they need for patient management. Imagine lavish reports, pre-filled with visual images of segmentations, graphs showing longitudinal changes, and quantitative metrics allowing for robust patient monitoring. These tools will become one of the many technological innovations that physicians depend on for diagnosis and treatment planning. I am also excited about the enthusiasm and support from the medical community. When I started, not too many clinicians were eager about computer-assisted image analysis, but the sentiment has shifted.
Are you working on any interesting projects right now?
In my lab, we concentrate on developing complete biomarker systems for medical imaging modalities using AI, with a special focus on neurological MRI for dementia, aging, vascular disease, and pediatrics, as well as digital pathology images for breast cancer, renal, and colon pathology. I am interested in improving efficiency and diagnostic accuracy for these clinical problems since they place high burdens on healthcare systems and patients – and would greatly benefit from automation. A major focus of the lab is knowledge translation, and creating systems that can be implemented in a variety of imaging centers. For this, we work with leading clinicians and companies to optimize translation opportunities. We also use AI tools to learn more about disease mechanisms to stratify patients in clinical trials, develop new therapies, and identify disease earlier.
What advice would you give to students following in your footsteps?
Do what you love! Engineering is surely not easy, and being driven and passionate about your work makes the journey fun and enjoyable. Every day I wake up excited about the possibilities and future which lets me stay focused and determined – even when the going gets tough! Consider every opportunity – as an opportunity to try new things, learn, explore, and challenge yourself. Hold yourself to a high standard, be accountable, and aim for the stars. Hard work pays off in the end and your efforts will be rewarded – don’t forget that!
